This is the fifth part on the evolution of human mating behavior, comparing evidence for promiscuity and pair-bonding in our species. Please see the introduction here.
“One of these days I will wake up – which I think I have done already – and realize to myself that I really do love. I find it very difficult to allow my whole life to rest on the existence of another creature. I find it equally difficult, because of my innate arrogance, to believe in the idea of love. There is no such thing, I say to myself. There is lust, of course, and usage, and jealousy, and desire and spent powers, but no such thing as the idiocy of love. Who invented that concept? I have wracked my shabby brains and can find no answer.” (Letter from Richard Burton to Elizabeth Taylor)
“Emotions and motivations (are) hierarchically ordered in the brain. Fear can overcome joy, for example. Jealousy can stifle tenderness…But in this pecking order of basic and complex emotions, background feelings and powerful drives, romantic love holds a special place: close to the zenith, the pinnacle, the top. Romantic love can dominate the drive to eat and sleep. It can stifle fear, anger, or disgust. It can override one’s sense of duty to family or friends. It can even triumph over the will to live. As Keats said, ‘I could die for you.’ ” (Helen Fisher 2004: 97-8)
“Only love can leave such a mark / But only love can heal such a scar.” (Paul Hewson)
Attempting to summarize the evolution of romantic love and pair-bonding in humans is an enormous task. This is exacerbated by many complications: the literature on these topics is vast; the term ‘pair-bond’ has been used in different ways by different primatologists in the past and is often erroneously confused with ‘monogamy’; the neurobiology of romantic love – which is not a prerequisite of pair-bonding – is complex; the question of when/if humans became pair-bonded in our evolutionary history is highly speculative; and then there is the complicating factor of culture. Furthermore, parts two through four in this series addressed ten biological traits indicating that humans have promiscuity built into us to some degree, and that monogamy, or at least lifelong monogamy, does not come easily.
Nonetheless, there is some very solid evidence that our species has some strong pair-bonding tendencies. This is not to say that pair-bonding and romantic love are automatically linked, but I group them here because it seems logical that pair-bonding would be facilitated by romantic love, at least in humans. Some anthropologists have argued that the ethnographic record shows that romantic love is almost universally found across human cultures, and that – contrary to Richard Burton’s armchair philosophizing – it is not an invented construct, but an innate biological attribute of our species. However, when reading the literature on this, one is also confronted by the reality that there is no guarantee that romantic love endures for life (in fact, it’s most intense phase seems to last about one-and-a-half to three years), or that pair-bonds are necessarily exclusive.
But let’s begin this post as cheerfully as possible. Romantic love, it seems, is a real thing, and may even be observed and measured operating in our highly complex, evolved brains. It has been described as one of the most “truly authentic and deeply moving experiences” that a person can ever have (Jankowiak and Paladino 2008: 25). Of course, love’s outcome is obviously affected by circumstances: whether it is returned or in vain, how it fits into cultural norms and social expectations, if it is opportune or ill-timed, as well as many other factors. For now, let’s be content that it exists at all, and explore why (with much speculation) this may have evolved.
Biological Correlates of Romantic Love
Not everyone uses the same vocabulary in defining erotic love, but most researchers make some demarcation between its more and less intense forms. Jankowiak and Paladino (2008: 2,18) distinguish between romantic love (also called passionate love, infatuation, or limerence) and comfort love (also known as companionship or attachment). Comfort love is not necessarily void of passion, but revolves more around long-term attachment, friendship, and deeply interwoven lives. By contrast, romantic love is much more intense, experienced earlier in a relationship, and involves the idealization of another person.
Helen Fisher and colleagues (2002:415-17) argued that romantic love includes a consistent suite of traits that cut across cultures. In a sample of Japanese and American respondents, they found thirteen characteristics that were reliably associated with intense romantic love, with few differences between the two samples. A partial list includes:
- obsessive, ‘intrusive thinking’
- thinking that the other person is unique
- prioritizing emotional ‘union’ over sexual desire
- focusing on positive qualities of the person, while overlooking negative ones
- increased energy and exhilaration
- a high sense of empathy and altruism toward the person
- sleeplessness and loss of appetite
- feeling greater connection to the person during adversity
- feeling that intense romantic love is ‘involuntary,’ but also temporary
To Fisher, one of the most important of these is intrusive thinking, as many respondents who were in love estimated that they thought about the other person more than 85% of the time. As one can imagine, this could be either a blessing or a curse, depending on circumstances (perhaps you don’t have to imagine if you’ve experienced this firsthand). Fisher compares the most intense stages of romantic love to an addiction – “a blissful dependency when one’s love is returned, a painful, sorrowful, and often destructive craving when one’s love is spurned” (2004: 53). The addiction analogy may not be too far off. Marazziti et al (1999) noted biological similarities between obsessive-compulsive disorder and romantic love, in that in both have low density of serotonin transporters and involve fixation on an ‘overvalued idea.’
Other biological factors are likely involved. In monogamous prairie voles (rodents of the genus Microtus), males and females form a pair-bond and remain in close proximity to each other for weeks to help raise offspring, while becoming territorial and aggressive toward unfamiliar individuals. Insel and Shapiro (1992) found that in comparison to their promiscuous montane vole cousins, who do not pair-bond, prairie vole brains have much denser levels of receptors for neuropeptides associated with affiliative behaviors (ex. oxytocin receptors in the prelimbic cortex and vasopressin receptors in the ventral pallidum). At least for prairie voles, pair-bonding seems to be stimulated by copulation itself, during which vasopressin and oxytocin (the so-called ‘love hormone’) are released.1 The effect is so predictable that experimentally blocking either vasopressin in prairie vole males or oxytocin in prairie vole females will prevent pair-bonding behavior (Insel and Young 2001).Injecting these directly into the central nervous system will have the opposite effect, and pair-bonding will occur irrespective of whether copulation has occurred. However, these effects are not seen in montane voles, due to the lower density of receptors.
Is that simple? Probably not. And in humans, for a drive as complex as romantic love, the biology is assuredly more involved than any single neurotransmitter or brain region. Much research has focused on the dopamine reward system and its connection to romantic love. In fMRI scans of individuals who reported being happily in love for an average of 7 months, several brain regions became active when looking at a photograph of their partner, including the right ventral tegmental area (VTA) and the caudate nucleus (Fisher et al 2006). Elsewhere, Fisher (2004:71) refers to the VTA as the “mother lode for dopamine making cells,” which – given its central location in the brain – can project dopamine to various regions associated with reward detection and expectation, such as the caudate nucleus. Bartels and Zeki (2000) confirmed elevated activity in the caudate nucleus in individuals in love, and also found activity in other brain regions that became active when using various drugs associated with euphoria, including cocaine. There is also some preliminary reporting that the feeling of romantic love can be triggered in the brain in less than a fifth of a second, suggesting that it is to some extent unconscious and involuntary.
At the risk of getting tedious, I will highlight one more neuro correlate of love because my guess is it may have some properties that are unique to humans. Fisher et al 2006 noted that the lateral orbitofrontal cortex, among other regions, was active in people who recently had been rejected in love. This is intriguing, because this area has also been associated with the capacity for a ‘theory of mind’ (understanding that others have mental states and intentions), an ability arguably limited to humans and possibly chimpanzees (Schmelz et al 2011). I emphasize this because while pair-bonding may be found in other species, I would think that the ability to understand others’ mental states at a sophisticated level would set human romantic love qualitatively apart. Pair-bonded prairie voles may very well spend lots of time together, but I seriously doubt they wonder how their beloved is feeling and thinking in a way that comes remotely close to what humans do. Perhaps that is one reason early onset romantic love is so intense in humans, with so much energy spent – seemingly against their will – on gauging how the person of their affection feels about them. Our mental and emotional connections, facilitated by language, occur at a level that prairie voles, swans, or gibbons can never approach. Nor do other species view sex or love as pathways to transcendence or attempt to convey their emotions, however feebly, by singing about rolling in the deep, giving up the ghost, fading into someone, la vie en rose, or (my personal favorite) floating into the mystic.
A Biocultural View
While there may be consistent neurobiological responses associated with romantic love (though this needs verification in more populations), it is essential to remember that almost every aspect of human behavior is affected by culture. Romantic love is no exception.2 Jankowiak and Paladino argue that romantic love is a near human universal, which “arises from forces within the hominid brain that are independent of the socially constructed mind” (2008: 9). That is a powerful statement. However, they and their co-authors spend an entire book noting the different ways various cultures interpret, experience, act upon, and prioritize romantic love. In their analysis, the most recent cross-cultural data indicate that at least some recognition of romantic love has been found in 91% (151 of 166) of sampled cultures, indicating its near ubiquity. (They provide a thorough appendix of sources for skeptics).
In some cultures, romantic love is idealized. However, in others it is conceptualized as madness or an illness requiring treatment. This is true of the Makassarese in Indonesia, the Taita in Kenya, and parts of Japan and Sri Lanka (Röttger-Rössler 2008: 172). Among the Lahu of southwest China, the cultural ideal is for husband and wife to work together harmoniously as farmers, and courtship songs center around this theme (Du 2008: 107). Romantic love is downplayed, public displays of sexual desire are extremely discouraged, and elopement and divorce are highly restricted. Du argues that the high demands of the cultural ideal, combined with social changes that followed Maoism, have proven too restrictive. In one small village cluster with a population of two thousand Lahu, 72 love suicides were recorded over a forty-six year period. The fact that this is recognizable to us, perhaps through the story of Romeo and Juliet (Radiohead version here), and not automatically dismissed as some exotic, circumscribed culture-bound syndrome, illustrates how we can empathize with how gut-wrenching it can be when culture demands the suppression of powerful emotions. The same could be said of any romantic or sexual desires that a culture considers lesser, illegitimate, or deviant. However, as individuals and societies mature and become more tolerant, it gets better.3
But the power of erotic or romantic emotions is enough to overwhelm even the most sensible. Even the erudite philosopher David Hume, who famously stated that “reason is, and ought only to be the slave of the passions,” succumbed by falling vainly in love with a married woman (Zaretsky 2011). Similarly, as a young man George Washington seemed to have been unable to erase the married Sally Fairfax from his mind (Fleming 2009). From Marie Curie, to Albert Einstein and Dmitri Mendeleev; history is rife with imprudent loves. Almost everybody plays the fool, from the brilliant to the not-so-brilliant. At least in this regard, nature appears to have gone overboard. Why would such a system evolve?
The Evolution of Pair-bonding
Fuentes (2002) wrote that, technically, ‘pair-bonding’ describes a social system of primates living in groups of two adults plus their offspring, while ‘monogamy’ refers solely to a mating pattern. The reason this is important is because there are likely quantitative and qualitative differences between a pair-bond in species living in two-adult groups (who have virtually no one else around) and a pair-bond in a larger, more social group. However, a general definition of a pair-bond might be “a predictable social and assumedly sexual relationship between two adults” (Fuentes p. 954). Still, questions remain. How predictable, enduring, or exclusive must the relationship be to be considered pair-bonded?
Fuentes listed many possible reasons for a species to be pair-bonded: females are too widely dispersed to allow any other mating option; mate-guarding (males prevent others from sexual access); females choose a male to act as a body guard against predators or other conspecific males; males are needed to prevent infanticide; or for male paternal care. To Bernard Chapais (2009), mate-guarding likely came first in our evolutionary history, which in effect acted as a pre-adaptation for biparental care later on. His reasoning for this is that there are some species of pair-bonded primates where mate guarding exists in the absence of any paternal care. However, when all of this occurred in humans is a matter of speculation.
Another common argument is that males and females would have formed a ‘sex contract’ at some point in our evolutionary history, trading female sexual fidelity in exchange for male parental care for our relatively helpless, altricial human infants. Sarah Hrdy (2009), however, outlines the holes with this idea, including the fact that ethnographic data show that hunter-gatherer fathers are not always obsessed with paternity and that communal alloparenting is common where paternal care is unavailable.
Chapais also argues that the transition from a more promiscuous mating system as seen in chimpanzees and bonobos to one with more ‘stable breeding bonds’ (he prefers this term to ‘pair-bonded’) was likely an indirect route, and had an intermediate polygynous stage. However, he feels that the development of more exclusive, stable breeding bonds was one of the most important transitions in human evolutionary history. In a very interesting hybrid hypothesis, he incorporates primatology, evolutionary theory, and Claude Levi-Strauss’ ideas on the importance of reciprocal exogamy. Marriage between groups would maintain ties between neighboring bands of foragers, facilitating social aggregation and the formation of tribes. In his estimation, this could not occur without stable breeding bonds, which would also make paternity, extended kin, and in-laws more recognizable. Again, the timing of this in evolutionary history is speculative, but he makes the point that reciprocal exogamy probably couldn’t occur without some mental capacity for trade and exchange. Instead of individuals trading objects, we have a scenario of groups ‘trading’ individuals in marriage. It’s tempting to speculate when this could have originated. In lab settings, capuchin monkeys and chimpanzees have been trained to exchange tokens for rewards (deWaal 2006), but this doesn’t seem to occur naturally. We might also look to the fossil record and the long-range transport of stone tools, which implies the existence of trade networks. Whereas chimpanzees rarely carry tools beyond a few hundred meters from their source of origin, archaeological data show that our ancestors were transporting raw materials for stone tools up to 100 km after about 1.0 million years ago, indicating a mental ability for trade and possibly language (Marwick 2003). There are details to be worked out, but these are interesting ideas nonetheless.
On a related note, chimpanzee mating behavior is not always simply promiscuous. They employ other behaviors as well, such as mate-guarding, where high-ranking males monopolize access to estrus females. Intriguingly, chimps also engage in consortships, where a single male and female leave the group and mate exclusively for a single estrus cycle (Mitani et al 2002). Mitani noted that while female chimpanzees often seem interested in exclusive mating, males sometimes coerce females to leave the group with them through “considerable aggression.” Genetic paternity tests confirm that all of the above are effective reproductive strategies. In one small study of twelve infants, 25% were conceived during consortships. These bonds are not nearly enduring enough to fit Chapais’ concept of stable breeding bonds, but Meredith Small (1990) has written elsewhere that consortships could be the predecessor of something akin to a pair-bond.
How Long Do Pair Bonds Last?
The duration of pair-bonds, where they exist, surely varies across primate species. It’s interesting that most descriptions in primatology, such as Chapais’ ‘stable breeding bonds,’ leave a lot of wiggle room. Quinlan wrote that pair-bonds are ‘relatively enduring’ (2008: 227). Fuentes (2002) defined a sexual pair-bond as lasting more than a year. But nobody says that pair-bonds last forever. In fact, no primate species mates for life, and as seen in part two, that includes humans as well. Speculating on pair-bonds in early hominin evolution 3.5 million years ago, Helen Fisher wrote that “I don’t see why these primordial pair-bonds needed to be permanent” (2004: 132).
To Fisher, romantic love evolved early in human evolution for its “timeless purpose of weeding out unsuitable mates, focusing one’s attention on a ‘special’ other, forming a visible pair-bond with this beloved, and remaining sexually faithful to ‘him’ or ‘her’ at least long enough to conceive a child together” (2004: 215). That qualifier at the end (“at least long enough to conceive a child together”) is an important one, suggesting that the intense emotional push that accompanies romantic love evolved to force a person to go beyond mere rational calculation in order to emotionally commit to a mate, likely for reproductive purposes. However, elsewhere Fisher states that humans evolved to be a serially pair-bonded species, based on the observation that most divorces in the world occur in the fourth year of marriage, which would at least allow a couple not only to conceive a child, but to raise one through the hazardous period of infancy (Fisher 1989). Still elsewhere, she argues that our unique life history strategy and delayed maturation would likely have led to long-term pair-bonds that lasted through childhood and adolescence to raise offspring (2004: 142). Which is it? I don’t know. Let’s agree that for human infants, children, juveniles, and adolescents, more parenting is probably better than less. We are highly dependent on others for food, protection, and socialization in these stages, with autonomy increasing with age. However, going back to Hrdy’s (2009) position, there is no guarantee that parental care has to come from the biological father, though it would obviously help.
But I think we may be dealing with proximate and ultimate causes too casually. As mentioned before in part two, sex and reproduction overlap, but not perfectly so. The same can be said about romantic love. Fisher acknowledges as much, noting that love can strike at any age, whether or not reproduction is even a viable biological option. Certainly, GLBT individuals can fall in love just as deeply as any heterosexual couple interested in procreation. Jesse Bering has written eloquently about this.
Long-term, even lifelong, pair-bonds are possible, but the biology suggests that they require effort, as they do not run on the intense neurobiological changes seen in early onset romantic love. Autopilot only lasts for so long. In yet another biomarker associated with romantic love, Emanuele et al (2006) found that peripheral levels of nerve growth factor, a neurotrophin that assists in neuron survival and maturation, was moderately correlated with one’s ‘passionate love score,’ ascertained by questionnaire (r = 0.34; p = 0.007). NGF has also been associated with increased levels of vasopressin, the neuropeptide described above in prairie voles. Levels of NGF were significantly higher in subjects who had fallen in love in the last 6 months than two control groups: single people and people who had been in a longer-term relationship for an average of 4 years. At follow-up, after the ‘in love’ group had been in their relationship for another 12 to 24 months, both passionate love scores and NGF levels decreased significantly (from 227 to 125 pg/ml; p = 0.001) and became indistinguishable from the control groups. The authors summarized their findings as suggesting that: “in humans the neurochemical bases of early-stage romantic love may be substantially different from those of longer term romantic relationships,” with the most intense stages likely lasting somewhere between 18 to 36 months (p. 293). Similarly, Marazziti et al (1999) found that the most intense effects of romantic love last about 12 to 18 months. This is consistent with the notion that intense early romantic love would at least give a pair-bonded couple enough time to conceive, but it would certainly be insufficient by itself to raise a child to reproductive maturity, or even through infancy. Other factors are likely involved as well.
However, as many researchers note, it is too easy to oversimplify the biology of romantic love and infer that once the intense initial spark has faded, that everything that follows is second-rate. Maybe. If you’re into drama. But once the crazy obsessing ebbs, and the neurotrophin, dopamine, and serotonin levels return to normal, companion love may take its place. And different types of love have their own neurological profiles and benefits (Ortigue et al 2010).
p.s. Bonus points if you know who Paul Hewson is, or if you counted the number of song references. Love is still a mystery to me, but I am trying hard to understand.…
- Part 1. Introduction Link
- Part 2. Promiscuity Link
- Part 3. Promiscuity (Genetics) Link
- Part 4. Promiscuity (Anatomy/Physiology) Link
- Part 5. Pair-Bonding and Romantic Love Link
- Part 6. Many Intimate Relationships Link
- Part 7. Is It Possible to Love Two People? Link
- Part 8. Love and Suffering Link
- Part 9. Love Is an Evolutionary Compromise Link
- Part 10. Wondrously Complex Paleo-Sex Link
1.) See Ed Yong’s piece on why ‘love-hormone’ is a misnomer for oxytocin.
2). In their book Medical Anthropology and the World System, Baer et al 2004 pointed out that even drunken behavior is culturally mediated, and is not solely the influence of the effects of alcohol metabolism.
3). Whenever there was a story on suicide in the news, my father would say that suicide is a ‘permanent solution to a temporary problem.’ He wasn’t ignoring the very real existence of depression or severe mental anguish, only that suicide offered no opportunity for second guessing.
Baer HA, Singer M, Susser I. 2003. Medical Anthropology and the World System. Greenwood. (Link)
Chapais B. 2009. The deep structure of human society: primate origins and evolution. In: Mind the Gap: Tracing the Origins of Human Universals. Springer pp 19-51 (Link)
De Waal F. 2006. Primates and Philosophers: How Morality Evolved. Princeton.
Du S. 2008. With one word and one strength: intimacy among the Lahu of Southwest China. In W Jankowiak (ed): Intimacies: Love & Sex Across Cultures. Pp. 95-121. Columbia Univ Press. (Link)
Emanuele E, Politi P, Bianchi M, Minoretti P, Bertona M, Geroldi D. 2006. Raised plasma nerve growth factor levels associated with early-stage romantic love. Psychoneuroendocrinology. 31(3):288-94. (Link)
Fisher H. 1989. Evolution of human serial pairbonding. American Journal of Physical Anthropology. 78(3):331–54. (Link)
Fisher H. 2004. Why We Love: The Nature and Chemistry of Romantic Love. Putnam. (Link)
Fisher HE, Aron A, Mashek D, Li H, Brown LL. 2002. Defining the brains systems of lust, romantic attraction, and attachement. Archives of Sexual Behavior 31(5):149-55. (Link)
Fleming T. 2009. George Washington in Love. American Heritage 59(3). (Link)
Fuentes A. 2002.Patterns and trends in primate pair bonds. International Journal of Primatology.23(5): 953-78. (Link)
Hrdy SB. 2009. Mothers and Others: The Evolutionary Origins of Mutual Understanding. Belknap. (Link)
Insel T, Shapiro LE. 1992. Oxytocin receptor distribution reflects social organization in monogamous and polygamous voles. Proc Natl Acad Sci U S A. 89(13): 5981–5. (Link)
Insel TR, Young LJ. 2001. The neurobiology of attachment. Nature Reviews. Neuroscience. 2(2):129-36. (Link)
Jankowiak WR, Paladino T. 2008. Desiring sex, longing for love. In W Jankowiak (ed): Intimacies: Love & Sex Across Cultures. Pp. 1-36. Columbia Univ Press. (Link)
Kim HS, Sherman DK, Sasaki JY, et al. 2010. Culture, distress, and oxytocin receptor polymorphism (OXTR) interact to inﬂuence emotional support seeking. Proc Natl Acad Sci U S A. 107(36):15717-21. (Link)
Marazziti D, Akiskal HS, Rossi A, Cassano GB. 1999. Alteration of the platelet serotonin transporter in romantic love. Psychol Med. 29(3):741-5. (Link)
Marwick B. 2003. Pleistocene exchange networks as evidence for the evolution of language. Cambridge Archaeological Journal 13(1): 67–81. (Link)
Mitani JC, Watts DP, Muller MN. 2002. Recent developments in the study of wild chimpanzee behavior. Evolutionary Anthropology: Issues, News, and Reviews. 11 (1): 9–25.
Ortigue S, Bianchi-Demicheli F, Patel N, Frum C, Lewis JW. 2010. Neuroimaging of love: fMRI meta-analysis evidence toward new perspectives in sexual medicine. Journal of Sexual Medicine. 7(11):3541-52. (Link)
Quinlan RJ. 2008. Human pair-bonds: evolutionary functions, ecological variation, and adaptive development. Evolutionary Anthropology 17:227-38. (Link)
Röttger-Rössler B. 2008. Voiced intimacies: verbalized experiences of love and sexuality in an Indonesian society. In W Jankowiak (ed): Intimacies: Love & Sex Across Cultures. Pp. 148-73. Columbia Univ Press. (Link)
Schmelz M, Call J, Tomasello M. 2011. Chimpanzees know that others make inferences. Proc Natl Acad Sci USA. 108(7):3077-9. (Link)
Small M. 1990. Consortships and conceptions in captive rhesus macaques (Macaca mulatta). Primates 31(3): 339-50. (Link)
Zaretsky R. 2011. A philosopher in love. New York Times May 6. (Link)